*Assistant Professor, Department of Mechanical Engineering, Sri Krishna College of Engineering and Technology, Coimbatore, Tamilnadu, India
**Professor, Department of Mechanical Engineering, Sri Krishna College of Engineering and Technology, Coimbatore, Tamilnadu, India
***UG Scholar, Department of Mechanical Engineering, Sri Krishna College of Engineering and Technology, Coimbatore, Tamilnadu, India
Online published on 12 January, 2017.
The present work is to investigate the effect of significant process parameters on responses and to validate the developed ANN model with the experimental data. To achieve this, 30 number of experiments were carried out on Air Hardened Tool Steel (AISI A2) suspended in deionized water dielectric with copper as tool electrode by using full factorial central composite experimental design based Response Surface Methodology (RSM). The Air hardened tool steel is chosen as the specimen due to its wide range of applications such as in large blanking dies, master hubs, precision tools and coining dies. The various process parameters which can be varied such as Pulse on time, Pulse off time, Peak Current and Voltage are chosen as input parameters. For the selection of dominant process parameters, preliminary experiments were carried out. The output responses viz., Material Removal Rate, Tool Wear Rate, Taper Angle, Circularity and Perpendicularity were measured and analyzed using Design Expert software analytically as well as graphically in order to evaluate the performances of the EDM process. Regression model is developed for predicting Material Removal Rate, Tool Wear Rate, Taper Angle, Circularity and Perpendicularity, in terms of interactive and individual terms of machining parameters through RSM, utilizing relevant experimental results as obtained through experimentation. The research outcome identifies significant parameters and their effects on process performances on AISI A2 tool steel. The adequacy of the above proposed model is also tested using analysis of variance (ANOVA) method. Surface plot for the responses (Material removal rate, Tool wear rate, Taper angle Circularity and Perpendicularity) were obtained considering the significant parameters. The program is developed in MATLAB using Neural Networks based back propagation algorithm and the results are compared with the experimental data. A validated mathematical model will be useful for the Industrial applications to predict the responses without performing unnecessary experiments for AISI A2 tool steel. This leads to increase the performance and economic productivity of drilling EDM process.
Drilling EDM, AISI A2 steel, RSM, ANN, Circularity
